Upload folder using huggingface_hub
#1
by
krishnakalyan3
- opened
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +2113 -0
- training_args.bin +3 -0
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a43c4f875942463645533430d9c55b742442309a4afaf5ad89a154734af9de26
|
| 3 |
+
size 379087640
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f8bae18bbc9d80f9de7a5d0ee12c76712e6edb5d99f00ce4cda7f91c23205d1f
|
| 3 |
+
size 3152668
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:88670816ebb9d3eff703b52fe5798928887886c40c85d6efedd4d8474e1ad671
|
| 3 |
+
size 14308
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:561c3d7591491183cabf86f8f5d391ee2b7ccd9426d42b0a4e9cdb8d70d862c3
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.13265489532696664,
|
| 3 |
+
"best_model_checkpoint": "/workspace/disk2/krishna/checkpoints/checkpoint-1300",
|
| 4 |
+
"epoch": 0.13,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 1300,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.001,
|
| 13 |
+
"grad_norm": 1.7856025695800781,
|
| 14 |
+
"learning_rate": 5.000000000000001e-07,
|
| 15 |
+
"loss": 0.682,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.001,
|
| 20 |
+
"eval_cos_sim": 0.30435964465141296,
|
| 21 |
+
"eval_loss": 0.6965229106162732,
|
| 22 |
+
"eval_runtime": 181.358,
|
| 23 |
+
"eval_samples_per_second": 22.056,
|
| 24 |
+
"eval_steps_per_second": 0.347,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.002,
|
| 29 |
+
"grad_norm": 1.632306694984436,
|
| 30 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 31 |
+
"loss": 0.6902,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.002,
|
| 36 |
+
"eval_cos_sim": 0.3060647249221802,
|
| 37 |
+
"eval_loss": 0.6948180051063245,
|
| 38 |
+
"eval_runtime": 158.9409,
|
| 39 |
+
"eval_samples_per_second": 25.167,
|
| 40 |
+
"eval_steps_per_second": 0.396,
|
| 41 |
+
"step": 20
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.003,
|
| 45 |
+
"grad_norm": 1.4781064987182617,
|
| 46 |
+
"learning_rate": 1.5e-06,
|
| 47 |
+
"loss": 0.6879,
|
| 48 |
+
"step": 30
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.003,
|
| 52 |
+
"eval_cos_sim": 0.30893880128860474,
|
| 53 |
+
"eval_loss": 0.6919441828941053,
|
| 54 |
+
"eval_runtime": 163.5656,
|
| 55 |
+
"eval_samples_per_second": 24.455,
|
| 56 |
+
"eval_steps_per_second": 0.385,
|
| 57 |
+
"step": 30
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.004,
|
| 61 |
+
"grad_norm": 1.4996154308319092,
|
| 62 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 63 |
+
"loss": 0.6805,
|
| 64 |
+
"step": 40
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.004,
|
| 68 |
+
"eval_cos_sim": 0.3130052089691162,
|
| 69 |
+
"eval_loss": 0.6878782258247084,
|
| 70 |
+
"eval_runtime": 160.8804,
|
| 71 |
+
"eval_samples_per_second": 24.863,
|
| 72 |
+
"eval_steps_per_second": 0.392,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.005,
|
| 77 |
+
"grad_norm": 1.7979084253311157,
|
| 78 |
+
"learning_rate": 2.5e-06,
|
| 79 |
+
"loss": 0.6795,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.005,
|
| 84 |
+
"eval_cos_sim": 0.3181459903717041,
|
| 85 |
+
"eval_loss": 0.6827379918311781,
|
| 86 |
+
"eval_runtime": 159.4021,
|
| 87 |
+
"eval_samples_per_second": 25.094,
|
| 88 |
+
"eval_steps_per_second": 0.395,
|
| 89 |
+
"step": 50
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.006,
|
| 93 |
+
"grad_norm": 1.4049359560012817,
|
| 94 |
+
"learning_rate": 3e-06,
|
| 95 |
+
"loss": 0.6781,
|
| 96 |
+
"step": 60
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.006,
|
| 100 |
+
"eval_cos_sim": 0.32444360852241516,
|
| 101 |
+
"eval_loss": 0.6764405532096571,
|
| 102 |
+
"eval_runtime": 159.9515,
|
| 103 |
+
"eval_samples_per_second": 25.008,
|
| 104 |
+
"eval_steps_per_second": 0.394,
|
| 105 |
+
"step": 60
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.007,
|
| 109 |
+
"grad_norm": 1.7637789249420166,
|
| 110 |
+
"learning_rate": 3.5000000000000004e-06,
|
| 111 |
+
"loss": 0.6718,
|
| 112 |
+
"step": 70
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.007,
|
| 116 |
+
"eval_cos_sim": 0.331728994846344,
|
| 117 |
+
"eval_loss": 0.6691558842872327,
|
| 118 |
+
"eval_runtime": 160.8921,
|
| 119 |
+
"eval_samples_per_second": 24.861,
|
| 120 |
+
"eval_steps_per_second": 0.392,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.008,
|
| 125 |
+
"grad_norm": 1.7560211420059204,
|
| 126 |
+
"learning_rate": 4.000000000000001e-06,
|
| 127 |
+
"loss": 0.6551,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.008,
|
| 132 |
+
"eval_cos_sim": 0.3400111794471741,
|
| 133 |
+
"eval_loss": 0.6608743510459608,
|
| 134 |
+
"eval_runtime": 172.4109,
|
| 135 |
+
"eval_samples_per_second": 23.2,
|
| 136 |
+
"eval_steps_per_second": 0.365,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.009,
|
| 141 |
+
"grad_norm": 1.7577630281448364,
|
| 142 |
+
"learning_rate": 4.5e-06,
|
| 143 |
+
"loss": 0.6566,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.009,
|
| 148 |
+
"eval_cos_sim": 0.34923624992370605,
|
| 149 |
+
"eval_loss": 0.6516506343101209,
|
| 150 |
+
"eval_runtime": 161.459,
|
| 151 |
+
"eval_samples_per_second": 24.774,
|
| 152 |
+
"eval_steps_per_second": 0.39,
|
| 153 |
+
"step": 90
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.01,
|
| 157 |
+
"grad_norm": 1.7723366022109985,
|
| 158 |
+
"learning_rate": 5e-06,
|
| 159 |
+
"loss": 0.6332,
|
| 160 |
+
"step": 100
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.01,
|
| 164 |
+
"eval_cos_sim": 0.35936808586120605,
|
| 165 |
+
"eval_loss": 0.6415200128768629,
|
| 166 |
+
"eval_runtime": 158.9596,
|
| 167 |
+
"eval_samples_per_second": 25.164,
|
| 168 |
+
"eval_steps_per_second": 0.396,
|
| 169 |
+
"step": 100
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.011,
|
| 173 |
+
"grad_norm": 1.7818301916122437,
|
| 174 |
+
"learning_rate": 5.500000000000001e-06,
|
| 175 |
+
"loss": 0.6297,
|
| 176 |
+
"step": 110
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.011,
|
| 180 |
+
"eval_cos_sim": 0.3702617287635803,
|
| 181 |
+
"eval_loss": 0.6306272282813734,
|
| 182 |
+
"eval_runtime": 161.4739,
|
| 183 |
+
"eval_samples_per_second": 24.772,
|
| 184 |
+
"eval_steps_per_second": 0.39,
|
| 185 |
+
"step": 110
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.012,
|
| 189 |
+
"grad_norm": 1.4825223684310913,
|
| 190 |
+
"learning_rate": 6e-06,
|
| 191 |
+
"loss": 0.6473,
|
| 192 |
+
"step": 120
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.012,
|
| 196 |
+
"eval_cos_sim": 0.381907194852829,
|
| 197 |
+
"eval_loss": 0.6189834599708265,
|
| 198 |
+
"eval_runtime": 162.0142,
|
| 199 |
+
"eval_samples_per_second": 24.689,
|
| 200 |
+
"eval_steps_per_second": 0.389,
|
| 201 |
+
"step": 120
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.013,
|
| 205 |
+
"grad_norm": 1.7424395084381104,
|
| 206 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 207 |
+
"loss": 0.611,
|
| 208 |
+
"step": 130
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.013,
|
| 212 |
+
"eval_cos_sim": 0.3940889537334442,
|
| 213 |
+
"eval_loss": 0.6068024029945082,
|
| 214 |
+
"eval_runtime": 161.0467,
|
| 215 |
+
"eval_samples_per_second": 24.838,
|
| 216 |
+
"eval_steps_per_second": 0.391,
|
| 217 |
+
"step": 130
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.014,
|
| 221 |
+
"grad_norm": 1.7474108934402466,
|
| 222 |
+
"learning_rate": 7.000000000000001e-06,
|
| 223 |
+
"loss": 0.5932,
|
| 224 |
+
"step": 140
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.014,
|
| 228 |
+
"eval_cos_sim": 0.4073576033115387,
|
| 229 |
+
"eval_loss": 0.5935340962623304,
|
| 230 |
+
"eval_runtime": 161.0017,
|
| 231 |
+
"eval_samples_per_second": 24.844,
|
| 232 |
+
"eval_steps_per_second": 0.391,
|
| 233 |
+
"step": 140
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.015,
|
| 237 |
+
"grad_norm": 1.6058062314987183,
|
| 238 |
+
"learning_rate": 7.5e-06,
|
| 239 |
+
"loss": 0.5824,
|
| 240 |
+
"step": 150
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.015,
|
| 244 |
+
"eval_cos_sim": 0.42149588465690613,
|
| 245 |
+
"eval_loss": 0.5793958764289564,
|
| 246 |
+
"eval_runtime": 167.7429,
|
| 247 |
+
"eval_samples_per_second": 23.846,
|
| 248 |
+
"eval_steps_per_second": 0.376,
|
| 249 |
+
"step": 150
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.016,
|
| 253 |
+
"grad_norm": 1.5788137912750244,
|
| 254 |
+
"learning_rate": 8.000000000000001e-06,
|
| 255 |
+
"loss": 0.5634,
|
| 256 |
+
"step": 160
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.016,
|
| 260 |
+
"eval_cos_sim": 0.43618762493133545,
|
| 261 |
+
"eval_loss": 0.564703805944795,
|
| 262 |
+
"eval_runtime": 159.9946,
|
| 263 |
+
"eval_samples_per_second": 25.001,
|
| 264 |
+
"eval_steps_per_second": 0.394,
|
| 265 |
+
"step": 160
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.017,
|
| 269 |
+
"grad_norm": 1.6113280057907104,
|
| 270 |
+
"learning_rate": 8.500000000000002e-06,
|
| 271 |
+
"loss": 0.5519,
|
| 272 |
+
"step": 170
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.017,
|
| 276 |
+
"eval_cos_sim": 0.4508235454559326,
|
| 277 |
+
"eval_loss": 0.5500685830329604,
|
| 278 |
+
"eval_runtime": 157.8438,
|
| 279 |
+
"eval_samples_per_second": 25.342,
|
| 280 |
+
"eval_steps_per_second": 0.399,
|
| 281 |
+
"step": 170
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.018,
|
| 285 |
+
"grad_norm": 1.6156154870986938,
|
| 286 |
+
"learning_rate": 9e-06,
|
| 287 |
+
"loss": 0.5407,
|
| 288 |
+
"step": 180
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.018,
|
| 292 |
+
"eval_cos_sim": 0.46566617488861084,
|
| 293 |
+
"eval_loss": 0.5352267394279188,
|
| 294 |
+
"eval_runtime": 159.9444,
|
| 295 |
+
"eval_samples_per_second": 25.009,
|
| 296 |
+
"eval_steps_per_second": 0.394,
|
| 297 |
+
"step": 180
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.019,
|
| 301 |
+
"grad_norm": 1.5958607196807861,
|
| 302 |
+
"learning_rate": 9.5e-06,
|
| 303 |
+
"loss": 0.5117,
|
| 304 |
+
"step": 190
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.019,
|
| 308 |
+
"eval_cos_sim": 0.4806951582431793,
|
| 309 |
+
"eval_loss": 0.5202000065063185,
|
| 310 |
+
"eval_runtime": 157.8685,
|
| 311 |
+
"eval_samples_per_second": 25.338,
|
| 312 |
+
"eval_steps_per_second": 0.399,
|
| 313 |
+
"step": 190
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.02,
|
| 317 |
+
"grad_norm": 1.6012651920318604,
|
| 318 |
+
"learning_rate": 1e-05,
|
| 319 |
+
"loss": 0.5116,
|
| 320 |
+
"step": 200
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.02,
|
| 324 |
+
"eval_cos_sim": 0.49558621644973755,
|
| 325 |
+
"eval_loss": 0.5053114848350233,
|
| 326 |
+
"eval_runtime": 155.395,
|
| 327 |
+
"eval_samples_per_second": 25.741,
|
| 328 |
+
"eval_steps_per_second": 0.405,
|
| 329 |
+
"step": 200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.021,
|
| 333 |
+
"grad_norm": 1.7965919971466064,
|
| 334 |
+
"learning_rate": 1.05e-05,
|
| 335 |
+
"loss": 0.4782,
|
| 336 |
+
"step": 210
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.021,
|
| 340 |
+
"eval_cos_sim": 0.5108910202980042,
|
| 341 |
+
"eval_loss": 0.49000780775296154,
|
| 342 |
+
"eval_runtime": 159.6303,
|
| 343 |
+
"eval_samples_per_second": 25.058,
|
| 344 |
+
"eval_steps_per_second": 0.395,
|
| 345 |
+
"step": 210
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.022,
|
| 349 |
+
"grad_norm": 1.791284441947937,
|
| 350 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 351 |
+
"loss": 0.4802,
|
| 352 |
+
"step": 220
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.022,
|
| 356 |
+
"eval_cos_sim": 0.5271998643875122,
|
| 357 |
+
"eval_loss": 0.4737012028907484,
|
| 358 |
+
"eval_runtime": 169.963,
|
| 359 |
+
"eval_samples_per_second": 23.535,
|
| 360 |
+
"eval_steps_per_second": 0.371,
|
| 361 |
+
"step": 220
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.023,
|
| 365 |
+
"grad_norm": 1.7923716306686401,
|
| 366 |
+
"learning_rate": 1.1500000000000002e-05,
|
| 367 |
+
"loss": 0.4513,
|
| 368 |
+
"step": 230
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.023,
|
| 372 |
+
"eval_cos_sim": 0.5435076355934143,
|
| 373 |
+
"eval_loss": 0.45739728691327036,
|
| 374 |
+
"eval_runtime": 160.753,
|
| 375 |
+
"eval_samples_per_second": 24.883,
|
| 376 |
+
"eval_steps_per_second": 0.392,
|
| 377 |
+
"step": 230
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.024,
|
| 381 |
+
"grad_norm": 1.7997322082519531,
|
| 382 |
+
"learning_rate": 1.2e-05,
|
| 383 |
+
"loss": 0.4288,
|
| 384 |
+
"step": 240
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.024,
|
| 388 |
+
"eval_cos_sim": 0.5595658421516418,
|
| 389 |
+
"eval_loss": 0.44134216930615366,
|
| 390 |
+
"eval_runtime": 163.2069,
|
| 391 |
+
"eval_samples_per_second": 24.509,
|
| 392 |
+
"eval_steps_per_second": 0.386,
|
| 393 |
+
"step": 240
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.025,
|
| 397 |
+
"grad_norm": 1.7595293521881104,
|
| 398 |
+
"learning_rate": 1.25e-05,
|
| 399 |
+
"loss": 0.4365,
|
| 400 |
+
"step": 250
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.025,
|
| 404 |
+
"eval_cos_sim": 0.5746485590934753,
|
| 405 |
+
"eval_loss": 0.4262613997672743,
|
| 406 |
+
"eval_runtime": 160.8892,
|
| 407 |
+
"eval_samples_per_second": 24.862,
|
| 408 |
+
"eval_steps_per_second": 0.392,
|
| 409 |
+
"step": 250
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.026,
|
| 413 |
+
"grad_norm": 1.614985704421997,
|
| 414 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 415 |
+
"loss": 0.4007,
|
| 416 |
+
"step": 260
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.026,
|
| 420 |
+
"eval_cos_sim": 0.5887113809585571,
|
| 421 |
+
"eval_loss": 0.4121987219070143,
|
| 422 |
+
"eval_runtime": 157.6029,
|
| 423 |
+
"eval_samples_per_second": 25.38,
|
| 424 |
+
"eval_steps_per_second": 0.4,
|
| 425 |
+
"step": 260
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.027,
|
| 429 |
+
"grad_norm": 1.6005358695983887,
|
| 430 |
+
"learning_rate": 1.3500000000000001e-05,
|
| 431 |
+
"loss": 0.4048,
|
| 432 |
+
"step": 270
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.027,
|
| 436 |
+
"eval_cos_sim": 0.6018706560134888,
|
| 437 |
+
"eval_loss": 0.39904179241406385,
|
| 438 |
+
"eval_runtime": 163.2521,
|
| 439 |
+
"eval_samples_per_second": 24.502,
|
| 440 |
+
"eval_steps_per_second": 0.386,
|
| 441 |
+
"step": 270
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.028,
|
| 445 |
+
"grad_norm": 1.7216198444366455,
|
| 446 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 447 |
+
"loss": 0.3793,
|
| 448 |
+
"step": 280
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.028,
|
| 452 |
+
"eval_cos_sim": 0.6152481436729431,
|
| 453 |
+
"eval_loss": 0.385668321153993,
|
| 454 |
+
"eval_runtime": 163.9341,
|
| 455 |
+
"eval_samples_per_second": 24.4,
|
| 456 |
+
"eval_steps_per_second": 0.384,
|
| 457 |
+
"step": 280
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.029,
|
| 461 |
+
"grad_norm": 1.7449027299880981,
|
| 462 |
+
"learning_rate": 1.45e-05,
|
| 463 |
+
"loss": 0.3614,
|
| 464 |
+
"step": 290
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.029,
|
| 468 |
+
"eval_cos_sim": 0.6277655959129333,
|
| 469 |
+
"eval_loss": 0.3731525359367079,
|
| 470 |
+
"eval_runtime": 170.8176,
|
| 471 |
+
"eval_samples_per_second": 23.417,
|
| 472 |
+
"eval_steps_per_second": 0.369,
|
| 473 |
+
"step": 290
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.03,
|
| 477 |
+
"grad_norm": 1.5057954788208008,
|
| 478 |
+
"learning_rate": 1.5e-05,
|
| 479 |
+
"loss": 0.3557,
|
| 480 |
+
"step": 300
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.03,
|
| 484 |
+
"eval_cos_sim": 0.6390805244445801,
|
| 485 |
+
"eval_loss": 0.36183924534069956,
|
| 486 |
+
"eval_runtime": 160.3353,
|
| 487 |
+
"eval_samples_per_second": 24.948,
|
| 488 |
+
"eval_steps_per_second": 0.393,
|
| 489 |
+
"step": 300
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.031,
|
| 493 |
+
"grad_norm": 1.5245262384414673,
|
| 494 |
+
"learning_rate": 1.55e-05,
|
| 495 |
+
"loss": 0.3729,
|
| 496 |
+
"step": 310
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.031,
|
| 500 |
+
"eval_cos_sim": 0.6503457427024841,
|
| 501 |
+
"eval_loss": 0.3505764603828138,
|
| 502 |
+
"eval_runtime": 161.447,
|
| 503 |
+
"eval_samples_per_second": 24.776,
|
| 504 |
+
"eval_steps_per_second": 0.39,
|
| 505 |
+
"step": 310
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.032,
|
| 509 |
+
"grad_norm": 1.4508600234985352,
|
| 510 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 511 |
+
"loss": 0.3476,
|
| 512 |
+
"step": 320
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.032,
|
| 516 |
+
"eval_cos_sim": 0.664078414440155,
|
| 517 |
+
"eval_loss": 0.33684703972088753,
|
| 518 |
+
"eval_runtime": 162.7312,
|
| 519 |
+
"eval_samples_per_second": 24.58,
|
| 520 |
+
"eval_steps_per_second": 0.387,
|
| 521 |
+
"step": 320
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.033,
|
| 525 |
+
"grad_norm": 1.545847773551941,
|
| 526 |
+
"learning_rate": 1.65e-05,
|
| 527 |
+
"loss": 0.3335,
|
| 528 |
+
"step": 330
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.033,
|
| 532 |
+
"eval_cos_sim": 0.6734265089035034,
|
| 533 |
+
"eval_loss": 0.32750138094174325,
|
| 534 |
+
"eval_runtime": 162.4883,
|
| 535 |
+
"eval_samples_per_second": 24.617,
|
| 536 |
+
"eval_steps_per_second": 0.388,
|
| 537 |
+
"step": 330
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.034,
|
| 541 |
+
"grad_norm": 1.5199618339538574,
|
| 542 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 543 |
+
"loss": 0.3369,
|
| 544 |
+
"step": 340
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 0.034,
|
| 548 |
+
"eval_cos_sim": 0.6810038685798645,
|
| 549 |
+
"eval_loss": 0.31992570402371345,
|
| 550 |
+
"eval_runtime": 161.4031,
|
| 551 |
+
"eval_samples_per_second": 24.783,
|
| 552 |
+
"eval_steps_per_second": 0.39,
|
| 553 |
+
"step": 340
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.035,
|
| 557 |
+
"grad_norm": 1.2642650604248047,
|
| 558 |
+
"learning_rate": 1.75e-05,
|
| 559 |
+
"loss": 0.3396,
|
| 560 |
+
"step": 350
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.035,
|
| 564 |
+
"eval_cos_sim": 0.6911267638206482,
|
| 565 |
+
"eval_loss": 0.3098063874458021,
|
| 566 |
+
"eval_runtime": 165.9265,
|
| 567 |
+
"eval_samples_per_second": 24.107,
|
| 568 |
+
"eval_steps_per_second": 0.38,
|
| 569 |
+
"step": 350
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.036,
|
| 573 |
+
"grad_norm": 1.2891690731048584,
|
| 574 |
+
"learning_rate": 1.8e-05,
|
| 575 |
+
"loss": 0.3015,
|
| 576 |
+
"step": 360
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.036,
|
| 580 |
+
"eval_cos_sim": 0.6987318992614746,
|
| 581 |
+
"eval_loss": 0.3022033410285658,
|
| 582 |
+
"eval_runtime": 173.6837,
|
| 583 |
+
"eval_samples_per_second": 23.03,
|
| 584 |
+
"eval_steps_per_second": 0.363,
|
| 585 |
+
"step": 360
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.037,
|
| 589 |
+
"grad_norm": 1.2396866083145142,
|
| 590 |
+
"learning_rate": 1.85e-05,
|
| 591 |
+
"loss": 0.3077,
|
| 592 |
+
"step": 370
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.037,
|
| 596 |
+
"eval_cos_sim": 0.7082959413528442,
|
| 597 |
+
"eval_loss": 0.2926409912322706,
|
| 598 |
+
"eval_runtime": 158.0401,
|
| 599 |
+
"eval_samples_per_second": 25.31,
|
| 600 |
+
"eval_steps_per_second": 0.399,
|
| 601 |
+
"step": 370
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.038,
|
| 605 |
+
"grad_norm": 1.2055168151855469,
|
| 606 |
+
"learning_rate": 1.9e-05,
|
| 607 |
+
"loss": 0.2895,
|
| 608 |
+
"step": 380
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.038,
|
| 612 |
+
"eval_cos_sim": 0.7168179750442505,
|
| 613 |
+
"eval_loss": 0.284124677202577,
|
| 614 |
+
"eval_runtime": 160.1742,
|
| 615 |
+
"eval_samples_per_second": 24.973,
|
| 616 |
+
"eval_steps_per_second": 0.393,
|
| 617 |
+
"step": 380
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.039,
|
| 621 |
+
"grad_norm": 1.1191834211349487,
|
| 622 |
+
"learning_rate": 1.9500000000000003e-05,
|
| 623 |
+
"loss": 0.2944,
|
| 624 |
+
"step": 390
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.039,
|
| 628 |
+
"eval_cos_sim": 0.7234470248222351,
|
| 629 |
+
"eval_loss": 0.2775011363242811,
|
| 630 |
+
"eval_runtime": 165.4196,
|
| 631 |
+
"eval_samples_per_second": 24.181,
|
| 632 |
+
"eval_steps_per_second": 0.381,
|
| 633 |
+
"step": 390
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.04,
|
| 637 |
+
"grad_norm": 1.097105860710144,
|
| 638 |
+
"learning_rate": 2e-05,
|
| 639 |
+
"loss": 0.2898,
|
| 640 |
+
"step": 400
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.04,
|
| 644 |
+
"eval_cos_sim": 0.7314009666442871,
|
| 645 |
+
"eval_loss": 0.2695479884360975,
|
| 646 |
+
"eval_runtime": 159.5742,
|
| 647 |
+
"eval_samples_per_second": 25.067,
|
| 648 |
+
"eval_steps_per_second": 0.395,
|
| 649 |
+
"step": 400
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.041,
|
| 653 |
+
"grad_norm": 1.040711760520935,
|
| 654 |
+
"learning_rate": 2.05e-05,
|
| 655 |
+
"loss": 0.2573,
|
| 656 |
+
"step": 410
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.041,
|
| 660 |
+
"eval_cos_sim": 0.7389957904815674,
|
| 661 |
+
"eval_loss": 0.26195128538357676,
|
| 662 |
+
"eval_runtime": 160.3231,
|
| 663 |
+
"eval_samples_per_second": 24.95,
|
| 664 |
+
"eval_steps_per_second": 0.393,
|
| 665 |
+
"step": 410
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.042,
|
| 669 |
+
"grad_norm": 1.0322086811065674,
|
| 670 |
+
"learning_rate": 2.1e-05,
|
| 671 |
+
"loss": 0.241,
|
| 672 |
+
"step": 420
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 0.042,
|
| 676 |
+
"eval_cos_sim": 0.7471259236335754,
|
| 677 |
+
"eval_loss": 0.25382448627697884,
|
| 678 |
+
"eval_runtime": 161.5365,
|
| 679 |
+
"eval_samples_per_second": 24.762,
|
| 680 |
+
"eval_steps_per_second": 0.39,
|
| 681 |
+
"step": 420
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.043,
|
| 685 |
+
"grad_norm": 0.9718736410140991,
|
| 686 |
+
"learning_rate": 2.15e-05,
|
| 687 |
+
"loss": 0.26,
|
| 688 |
+
"step": 430
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.043,
|
| 692 |
+
"eval_cos_sim": 0.7534591555595398,
|
| 693 |
+
"eval_loss": 0.24748690702664317,
|
| 694 |
+
"eval_runtime": 169.2487,
|
| 695 |
+
"eval_samples_per_second": 23.634,
|
| 696 |
+
"eval_steps_per_second": 0.372,
|
| 697 |
+
"step": 430
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.044,
|
| 701 |
+
"grad_norm": 0.8780993223190308,
|
| 702 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 703 |
+
"loss": 0.2255,
|
| 704 |
+
"step": 440
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.044,
|
| 708 |
+
"eval_cos_sim": 0.7604101896286011,
|
| 709 |
+
"eval_loss": 0.2405370173667616,
|
| 710 |
+
"eval_runtime": 161.7354,
|
| 711 |
+
"eval_samples_per_second": 24.732,
|
| 712 |
+
"eval_steps_per_second": 0.39,
|
| 713 |
+
"step": 440
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 0.045,
|
| 717 |
+
"grad_norm": 0.8193601369857788,
|
| 718 |
+
"learning_rate": 2.25e-05,
|
| 719 |
+
"loss": 0.2182,
|
| 720 |
+
"step": 450
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 0.045,
|
| 724 |
+
"eval_cos_sim": 0.7673761248588562,
|
| 725 |
+
"eval_loss": 0.2335676951621718,
|
| 726 |
+
"eval_runtime": 159.2952,
|
| 727 |
+
"eval_samples_per_second": 25.111,
|
| 728 |
+
"eval_steps_per_second": 0.395,
|
| 729 |
+
"step": 450
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.046,
|
| 733 |
+
"grad_norm": 0.8820355534553528,
|
| 734 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 735 |
+
"loss": 0.2153,
|
| 736 |
+
"step": 460
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.046,
|
| 740 |
+
"eval_cos_sim": 0.7723870277404785,
|
| 741 |
+
"eval_loss": 0.228554349443788,
|
| 742 |
+
"eval_runtime": 161.8756,
|
| 743 |
+
"eval_samples_per_second": 24.71,
|
| 744 |
+
"eval_steps_per_second": 0.389,
|
| 745 |
+
"step": 460
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.047,
|
| 749 |
+
"grad_norm": 0.9045171141624451,
|
| 750 |
+
"learning_rate": 2.35e-05,
|
| 751 |
+
"loss": 0.2245,
|
| 752 |
+
"step": 470
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.047,
|
| 756 |
+
"eval_cos_sim": 0.7763780951499939,
|
| 757 |
+
"eval_loss": 0.22456564476239152,
|
| 758 |
+
"eval_runtime": 160.6855,
|
| 759 |
+
"eval_samples_per_second": 24.893,
|
| 760 |
+
"eval_steps_per_second": 0.392,
|
| 761 |
+
"step": 470
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.048,
|
| 765 |
+
"grad_norm": 0.7304402589797974,
|
| 766 |
+
"learning_rate": 2.4e-05,
|
| 767 |
+
"loss": 0.2069,
|
| 768 |
+
"step": 480
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 0.048,
|
| 772 |
+
"eval_cos_sim": 0.7830195426940918,
|
| 773 |
+
"eval_loss": 0.21792557956921524,
|
| 774 |
+
"eval_runtime": 162.6536,
|
| 775 |
+
"eval_samples_per_second": 24.592,
|
| 776 |
+
"eval_steps_per_second": 0.387,
|
| 777 |
+
"step": 480
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 0.049,
|
| 781 |
+
"grad_norm": 0.6550305485725403,
|
| 782 |
+
"learning_rate": 2.45e-05,
|
| 783 |
+
"loss": 0.2001,
|
| 784 |
+
"step": 490
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 0.049,
|
| 788 |
+
"eval_cos_sim": 0.7868452072143555,
|
| 789 |
+
"eval_loss": 0.21410264875637955,
|
| 790 |
+
"eval_runtime": 163.12,
|
| 791 |
+
"eval_samples_per_second": 24.522,
|
| 792 |
+
"eval_steps_per_second": 0.386,
|
| 793 |
+
"step": 490
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.05,
|
| 797 |
+
"grad_norm": 0.6369508504867554,
|
| 798 |
+
"learning_rate": 2.5e-05,
|
| 799 |
+
"loss": 0.1935,
|
| 800 |
+
"step": 500
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.05,
|
| 804 |
+
"eval_cos_sim": 0.7914958596229553,
|
| 805 |
+
"eval_loss": 0.2094592919563002,
|
| 806 |
+
"eval_runtime": 164.3494,
|
| 807 |
+
"eval_samples_per_second": 24.338,
|
| 808 |
+
"eval_steps_per_second": 0.383,
|
| 809 |
+
"step": 500
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.051,
|
| 813 |
+
"grad_norm": 0.6336702108383179,
|
| 814 |
+
"learning_rate": 2.5500000000000003e-05,
|
| 815 |
+
"loss": 0.2018,
|
| 816 |
+
"step": 510
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 0.051,
|
| 820 |
+
"eval_cos_sim": 0.7958762645721436,
|
| 821 |
+
"eval_loss": 0.20508341314541764,
|
| 822 |
+
"eval_runtime": 163.0667,
|
| 823 |
+
"eval_samples_per_second": 24.53,
|
| 824 |
+
"eval_steps_per_second": 0.386,
|
| 825 |
+
"step": 510
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 0.052,
|
| 829 |
+
"grad_norm": 0.585249125957489,
|
| 830 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 831 |
+
"loss": 0.1854,
|
| 832 |
+
"step": 520
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 0.052,
|
| 836 |
+
"eval_cos_sim": 0.7985916137695312,
|
| 837 |
+
"eval_loss": 0.20236765434491105,
|
| 838 |
+
"eval_runtime": 161.7682,
|
| 839 |
+
"eval_samples_per_second": 24.727,
|
| 840 |
+
"eval_steps_per_second": 0.389,
|
| 841 |
+
"step": 520
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.053,
|
| 845 |
+
"grad_norm": 0.5437793731689453,
|
| 846 |
+
"learning_rate": 2.6500000000000004e-05,
|
| 847 |
+
"loss": 0.1795,
|
| 848 |
+
"step": 530
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.053,
|
| 852 |
+
"eval_cos_sim": 0.8020971417427063,
|
| 853 |
+
"eval_loss": 0.19886508896099991,
|
| 854 |
+
"eval_runtime": 164.1233,
|
| 855 |
+
"eval_samples_per_second": 24.372,
|
| 856 |
+
"eval_steps_per_second": 0.384,
|
| 857 |
+
"step": 530
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.054,
|
| 861 |
+
"grad_norm": 0.6123674511909485,
|
| 862 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 863 |
+
"loss": 0.1807,
|
| 864 |
+
"step": 540
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.054,
|
| 868 |
+
"eval_cos_sim": 0.8049777150154114,
|
| 869 |
+
"eval_loss": 0.19599416496502822,
|
| 870 |
+
"eval_runtime": 163.7538,
|
| 871 |
+
"eval_samples_per_second": 24.427,
|
| 872 |
+
"eval_steps_per_second": 0.385,
|
| 873 |
+
"step": 540
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.055,
|
| 877 |
+
"grad_norm": 0.5047230124473572,
|
| 878 |
+
"learning_rate": 2.7500000000000004e-05,
|
| 879 |
+
"loss": 0.1769,
|
| 880 |
+
"step": 550
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 0.055,
|
| 884 |
+
"eval_cos_sim": 0.8080028295516968,
|
| 885 |
+
"eval_loss": 0.19296009018170304,
|
| 886 |
+
"eval_runtime": 158.2319,
|
| 887 |
+
"eval_samples_per_second": 25.279,
|
| 888 |
+
"eval_steps_per_second": 0.398,
|
| 889 |
+
"step": 550
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 0.056,
|
| 893 |
+
"grad_norm": 0.4361515939235687,
|
| 894 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 895 |
+
"loss": 0.1783,
|
| 896 |
+
"step": 560
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 0.056,
|
| 900 |
+
"eval_cos_sim": 0.8107978701591492,
|
| 901 |
+
"eval_loss": 0.19017480518567031,
|
| 902 |
+
"eval_runtime": 164.6869,
|
| 903 |
+
"eval_samples_per_second": 24.289,
|
| 904 |
+
"eval_steps_per_second": 0.383,
|
| 905 |
+
"step": 560
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.057,
|
| 909 |
+
"grad_norm": 0.4613172113895416,
|
| 910 |
+
"learning_rate": 2.8499999999999998e-05,
|
| 911 |
+
"loss": 0.1714,
|
| 912 |
+
"step": 570
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.057,
|
| 916 |
+
"eval_cos_sim": 0.8138161301612854,
|
| 917 |
+
"eval_loss": 0.18716797020184464,
|
| 918 |
+
"eval_runtime": 167.1324,
|
| 919 |
+
"eval_samples_per_second": 23.933,
|
| 920 |
+
"eval_steps_per_second": 0.377,
|
| 921 |
+
"step": 570
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 0.058,
|
| 925 |
+
"grad_norm": 0.4287095069885254,
|
| 926 |
+
"learning_rate": 2.9e-05,
|
| 927 |
+
"loss": 0.1633,
|
| 928 |
+
"step": 580
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 0.058,
|
| 932 |
+
"eval_cos_sim": 0.8161399364471436,
|
| 933 |
+
"eval_loss": 0.1848432316993422,
|
| 934 |
+
"eval_runtime": 160.6945,
|
| 935 |
+
"eval_samples_per_second": 24.892,
|
| 936 |
+
"eval_steps_per_second": 0.392,
|
| 937 |
+
"step": 580
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 0.059,
|
| 941 |
+
"grad_norm": 0.442719429731369,
|
| 942 |
+
"learning_rate": 2.95e-05,
|
| 943 |
+
"loss": 0.1643,
|
| 944 |
+
"step": 590
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 0.059,
|
| 948 |
+
"eval_cos_sim": 0.8180400133132935,
|
| 949 |
+
"eval_loss": 0.1829425740455336,
|
| 950 |
+
"eval_runtime": 163.6004,
|
| 951 |
+
"eval_samples_per_second": 24.45,
|
| 952 |
+
"eval_steps_per_second": 0.385,
|
| 953 |
+
"step": 590
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.06,
|
| 957 |
+
"grad_norm": 0.38597649335861206,
|
| 958 |
+
"learning_rate": 3e-05,
|
| 959 |
+
"loss": 0.1645,
|
| 960 |
+
"step": 600
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 0.06,
|
| 964 |
+
"eval_cos_sim": 0.8204721808433533,
|
| 965 |
+
"eval_loss": 0.1805112529014296,
|
| 966 |
+
"eval_runtime": 158.8011,
|
| 967 |
+
"eval_samples_per_second": 25.189,
|
| 968 |
+
"eval_steps_per_second": 0.397,
|
| 969 |
+
"step": 600
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.061,
|
| 973 |
+
"grad_norm": 0.40187621116638184,
|
| 974 |
+
"learning_rate": 3.05e-05,
|
| 975 |
+
"loss": 0.1632,
|
| 976 |
+
"step": 610
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.061,
|
| 980 |
+
"eval_cos_sim": 0.8225390911102295,
|
| 981 |
+
"eval_loss": 0.17843791820752097,
|
| 982 |
+
"eval_runtime": 164.1328,
|
| 983 |
+
"eval_samples_per_second": 24.371,
|
| 984 |
+
"eval_steps_per_second": 0.384,
|
| 985 |
+
"step": 610
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 0.062,
|
| 989 |
+
"grad_norm": 0.36652931571006775,
|
| 990 |
+
"learning_rate": 3.1e-05,
|
| 991 |
+
"loss": 0.1587,
|
| 992 |
+
"step": 620
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 0.062,
|
| 996 |
+
"eval_cos_sim": 0.8244653344154358,
|
| 997 |
+
"eval_loss": 0.17650788643109275,
|
| 998 |
+
"eval_runtime": 160.7137,
|
| 999 |
+
"eval_samples_per_second": 24.889,
|
| 1000 |
+
"eval_steps_per_second": 0.392,
|
| 1001 |
+
"step": 620
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 0.063,
|
| 1005 |
+
"grad_norm": 0.39730513095855713,
|
| 1006 |
+
"learning_rate": 3.15e-05,
|
| 1007 |
+
"loss": 0.164,
|
| 1008 |
+
"step": 630
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 0.063,
|
| 1012 |
+
"eval_cos_sim": 0.8262067437171936,
|
| 1013 |
+
"eval_loss": 0.17477095749127342,
|
| 1014 |
+
"eval_runtime": 165.8251,
|
| 1015 |
+
"eval_samples_per_second": 24.122,
|
| 1016 |
+
"eval_steps_per_second": 0.38,
|
| 1017 |
+
"step": 630
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.064,
|
| 1021 |
+
"grad_norm": 0.37115684151649475,
|
| 1022 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 1023 |
+
"loss": 0.1562,
|
| 1024 |
+
"step": 640
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.064,
|
| 1028 |
+
"eval_cos_sim": 0.8285937905311584,
|
| 1029 |
+
"eval_loss": 0.17237986900555563,
|
| 1030 |
+
"eval_runtime": 166.1474,
|
| 1031 |
+
"eval_samples_per_second": 24.075,
|
| 1032 |
+
"eval_steps_per_second": 0.379,
|
| 1033 |
+
"step": 640
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 0.065,
|
| 1037 |
+
"grad_norm": 0.3438771367073059,
|
| 1038 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 1039 |
+
"loss": 0.1628,
|
| 1040 |
+
"step": 650
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 0.065,
|
| 1044 |
+
"eval_cos_sim": 0.8305408358573914,
|
| 1045 |
+
"eval_loss": 0.17043518688427878,
|
| 1046 |
+
"eval_runtime": 156.9198,
|
| 1047 |
+
"eval_samples_per_second": 25.491,
|
| 1048 |
+
"eval_steps_per_second": 0.401,
|
| 1049 |
+
"step": 650
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 0.066,
|
| 1053 |
+
"grad_norm": 0.445515513420105,
|
| 1054 |
+
"learning_rate": 3.3e-05,
|
| 1055 |
+
"loss": 0.1588,
|
| 1056 |
+
"step": 660
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 0.066,
|
| 1060 |
+
"eval_cos_sim": 0.8313919305801392,
|
| 1061 |
+
"eval_loss": 0.1695925650809951,
|
| 1062 |
+
"eval_runtime": 160.5009,
|
| 1063 |
+
"eval_samples_per_second": 24.922,
|
| 1064 |
+
"eval_steps_per_second": 0.393,
|
| 1065 |
+
"step": 660
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.067,
|
| 1069 |
+
"grad_norm": 0.30833250284194946,
|
| 1070 |
+
"learning_rate": 3.35e-05,
|
| 1071 |
+
"loss": 0.1597,
|
| 1072 |
+
"step": 670
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 0.067,
|
| 1076 |
+
"eval_cos_sim": 0.8327498435974121,
|
| 1077 |
+
"eval_loss": 0.16823711063610983,
|
| 1078 |
+
"eval_runtime": 163.7947,
|
| 1079 |
+
"eval_samples_per_second": 24.421,
|
| 1080 |
+
"eval_steps_per_second": 0.385,
|
| 1081 |
+
"step": 670
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.068,
|
| 1085 |
+
"grad_norm": 0.30146872997283936,
|
| 1086 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 1087 |
+
"loss": 0.1521,
|
| 1088 |
+
"step": 680
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.068,
|
| 1092 |
+
"eval_cos_sim": 0.8345705270767212,
|
| 1093 |
+
"eval_loss": 0.16642027666317893,
|
| 1094 |
+
"eval_runtime": 163.0767,
|
| 1095 |
+
"eval_samples_per_second": 24.528,
|
| 1096 |
+
"eval_steps_per_second": 0.386,
|
| 1097 |
+
"step": 680
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 0.069,
|
| 1101 |
+
"grad_norm": 0.3060350716114044,
|
| 1102 |
+
"learning_rate": 3.45e-05,
|
| 1103 |
+
"loss": 0.1499,
|
| 1104 |
+
"step": 690
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 0.069,
|
| 1108 |
+
"eval_cos_sim": 0.8357890844345093,
|
| 1109 |
+
"eval_loss": 0.16520177223431537,
|
| 1110 |
+
"eval_runtime": 161.8292,
|
| 1111 |
+
"eval_samples_per_second": 24.717,
|
| 1112 |
+
"eval_steps_per_second": 0.389,
|
| 1113 |
+
"step": 690
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 0.07,
|
| 1117 |
+
"grad_norm": 0.25746750831604004,
|
| 1118 |
+
"learning_rate": 3.5e-05,
|
| 1119 |
+
"loss": 0.1494,
|
| 1120 |
+
"step": 700
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 0.07,
|
| 1124 |
+
"eval_cos_sim": 0.8371787071228027,
|
| 1125 |
+
"eval_loss": 0.16381443121182396,
|
| 1126 |
+
"eval_runtime": 171.1282,
|
| 1127 |
+
"eval_samples_per_second": 23.374,
|
| 1128 |
+
"eval_steps_per_second": 0.368,
|
| 1129 |
+
"step": 700
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.071,
|
| 1133 |
+
"grad_norm": 0.27807655930519104,
|
| 1134 |
+
"learning_rate": 3.55e-05,
|
| 1135 |
+
"loss": 0.1503,
|
| 1136 |
+
"step": 710
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.071,
|
| 1140 |
+
"eval_cos_sim": 0.8385919332504272,
|
| 1141 |
+
"eval_loss": 0.16239957573162986,
|
| 1142 |
+
"eval_runtime": 173.6991,
|
| 1143 |
+
"eval_samples_per_second": 23.028,
|
| 1144 |
+
"eval_steps_per_second": 0.363,
|
| 1145 |
+
"step": 710
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 0.072,
|
| 1149 |
+
"grad_norm": 0.2822527587413788,
|
| 1150 |
+
"learning_rate": 3.6e-05,
|
| 1151 |
+
"loss": 0.1596,
|
| 1152 |
+
"step": 720
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 0.072,
|
| 1156 |
+
"eval_cos_sim": 0.8399993777275085,
|
| 1157 |
+
"eval_loss": 0.16099353077160786,
|
| 1158 |
+
"eval_runtime": 166.2107,
|
| 1159 |
+
"eval_samples_per_second": 24.066,
|
| 1160 |
+
"eval_steps_per_second": 0.379,
|
| 1161 |
+
"step": 720
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 0.073,
|
| 1165 |
+
"grad_norm": 0.3689976632595062,
|
| 1166 |
+
"learning_rate": 3.65e-05,
|
| 1167 |
+
"loss": 0.1522,
|
| 1168 |
+
"step": 730
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 0.073,
|
| 1172 |
+
"eval_cos_sim": 0.8407827615737915,
|
| 1173 |
+
"eval_loss": 0.16021420147168108,
|
| 1174 |
+
"eval_runtime": 163.1982,
|
| 1175 |
+
"eval_samples_per_second": 24.51,
|
| 1176 |
+
"eval_steps_per_second": 0.386,
|
| 1177 |
+
"step": 730
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 0.074,
|
| 1181 |
+
"grad_norm": 0.24420756101608276,
|
| 1182 |
+
"learning_rate": 3.7e-05,
|
| 1183 |
+
"loss": 0.1464,
|
| 1184 |
+
"step": 740
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 0.074,
|
| 1188 |
+
"eval_cos_sim": 0.8423337936401367,
|
| 1189 |
+
"eval_loss": 0.15866206934201188,
|
| 1190 |
+
"eval_runtime": 162.3325,
|
| 1191 |
+
"eval_samples_per_second": 24.641,
|
| 1192 |
+
"eval_steps_per_second": 0.388,
|
| 1193 |
+
"step": 740
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.075,
|
| 1197 |
+
"grad_norm": 0.2716599404811859,
|
| 1198 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 1199 |
+
"loss": 0.1458,
|
| 1200 |
+
"step": 750
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.075,
|
| 1204 |
+
"eval_cos_sim": 0.8430612683296204,
|
| 1205 |
+
"eval_loss": 0.157932794592256,
|
| 1206 |
+
"eval_runtime": 163.6706,
|
| 1207 |
+
"eval_samples_per_second": 24.439,
|
| 1208 |
+
"eval_steps_per_second": 0.385,
|
| 1209 |
+
"step": 750
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 0.076,
|
| 1213 |
+
"grad_norm": 0.27678823471069336,
|
| 1214 |
+
"learning_rate": 3.8e-05,
|
| 1215 |
+
"loss": 0.1446,
|
| 1216 |
+
"step": 760
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 0.076,
|
| 1220 |
+
"eval_cos_sim": 0.8438640236854553,
|
| 1221 |
+
"eval_loss": 0.15713081218945452,
|
| 1222 |
+
"eval_runtime": 159.6959,
|
| 1223 |
+
"eval_samples_per_second": 25.048,
|
| 1224 |
+
"eval_steps_per_second": 0.394,
|
| 1225 |
+
"step": 760
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 0.077,
|
| 1229 |
+
"grad_norm": 0.23866839706897736,
|
| 1230 |
+
"learning_rate": 3.85e-05,
|
| 1231 |
+
"loss": 0.1487,
|
| 1232 |
+
"step": 770
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 0.077,
|
| 1236 |
+
"eval_cos_sim": 0.8451653718948364,
|
| 1237 |
+
"eval_loss": 0.15582896663891738,
|
| 1238 |
+
"eval_runtime": 170.448,
|
| 1239 |
+
"eval_samples_per_second": 23.468,
|
| 1240 |
+
"eval_steps_per_second": 0.37,
|
| 1241 |
+
"step": 770
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.078,
|
| 1245 |
+
"grad_norm": 0.24985139071941376,
|
| 1246 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 1247 |
+
"loss": 0.143,
|
| 1248 |
+
"step": 780
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.078,
|
| 1252 |
+
"eval_cos_sim": 0.8460158705711365,
|
| 1253 |
+
"eval_loss": 0.15497446682202284,
|
| 1254 |
+
"eval_runtime": 168.3573,
|
| 1255 |
+
"eval_samples_per_second": 23.759,
|
| 1256 |
+
"eval_steps_per_second": 0.374,
|
| 1257 |
+
"step": 780
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 0.079,
|
| 1261 |
+
"grad_norm": 0.2435818761587143,
|
| 1262 |
+
"learning_rate": 3.9500000000000005e-05,
|
| 1263 |
+
"loss": 0.1464,
|
| 1264 |
+
"step": 790
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 0.079,
|
| 1268 |
+
"eval_cos_sim": 0.8470072150230408,
|
| 1269 |
+
"eval_loss": 0.15398560383068982,
|
| 1270 |
+
"eval_runtime": 163.7823,
|
| 1271 |
+
"eval_samples_per_second": 24.423,
|
| 1272 |
+
"eval_steps_per_second": 0.385,
|
| 1273 |
+
"step": 790
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 0.08,
|
| 1277 |
+
"grad_norm": 0.25132063031196594,
|
| 1278 |
+
"learning_rate": 4e-05,
|
| 1279 |
+
"loss": 0.1463,
|
| 1280 |
+
"step": 800
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 0.08,
|
| 1284 |
+
"eval_cos_sim": 0.8480408191680908,
|
| 1285 |
+
"eval_loss": 0.15294421245800913,
|
| 1286 |
+
"eval_runtime": 160.63,
|
| 1287 |
+
"eval_samples_per_second": 24.902,
|
| 1288 |
+
"eval_steps_per_second": 0.392,
|
| 1289 |
+
"step": 800
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 0.081,
|
| 1293 |
+
"grad_norm": 0.22080835700035095,
|
| 1294 |
+
"learning_rate": 4.05e-05,
|
| 1295 |
+
"loss": 0.1449,
|
| 1296 |
+
"step": 810
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 0.081,
|
| 1300 |
+
"eval_cos_sim": 0.8484798669815063,
|
| 1301 |
+
"eval_loss": 0.15251411964642467,
|
| 1302 |
+
"eval_runtime": 165.2445,
|
| 1303 |
+
"eval_samples_per_second": 24.207,
|
| 1304 |
+
"eval_steps_per_second": 0.381,
|
| 1305 |
+
"step": 810
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.082,
|
| 1309 |
+
"grad_norm": 0.21103359758853912,
|
| 1310 |
+
"learning_rate": 4.1e-05,
|
| 1311 |
+
"loss": 0.1425,
|
| 1312 |
+
"step": 820
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.082,
|
| 1316 |
+
"eval_cos_sim": 0.8491047024726868,
|
| 1317 |
+
"eval_loss": 0.15190097334134042,
|
| 1318 |
+
"eval_runtime": 165.2932,
|
| 1319 |
+
"eval_samples_per_second": 24.199,
|
| 1320 |
+
"eval_steps_per_second": 0.381,
|
| 1321 |
+
"step": 820
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 0.083,
|
| 1325 |
+
"grad_norm": 0.4737427234649658,
|
| 1326 |
+
"learning_rate": 4.15e-05,
|
| 1327 |
+
"loss": 0.1451,
|
| 1328 |
+
"step": 830
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 0.083,
|
| 1332 |
+
"eval_cos_sim": 0.8501515984535217,
|
| 1333 |
+
"eval_loss": 0.15085182239758432,
|
| 1334 |
+
"eval_runtime": 164.4705,
|
| 1335 |
+
"eval_samples_per_second": 24.32,
|
| 1336 |
+
"eval_steps_per_second": 0.383,
|
| 1337 |
+
"step": 830
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 0.084,
|
| 1341 |
+
"grad_norm": 0.19636596739292145,
|
| 1342 |
+
"learning_rate": 4.2e-05,
|
| 1343 |
+
"loss": 0.139,
|
| 1344 |
+
"step": 840
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 0.084,
|
| 1348 |
+
"eval_cos_sim": 0.8513444066047668,
|
| 1349 |
+
"eval_loss": 0.14965240146862924,
|
| 1350 |
+
"eval_runtime": 168.3369,
|
| 1351 |
+
"eval_samples_per_second": 23.762,
|
| 1352 |
+
"eval_steps_per_second": 0.374,
|
| 1353 |
+
"step": 840
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.085,
|
| 1357 |
+
"grad_norm": 0.22018375992774963,
|
| 1358 |
+
"learning_rate": 4.25e-05,
|
| 1359 |
+
"loss": 0.139,
|
| 1360 |
+
"step": 850
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.085,
|
| 1364 |
+
"eval_cos_sim": 0.8519036769866943,
|
| 1365 |
+
"eval_loss": 0.14909614183651865,
|
| 1366 |
+
"eval_runtime": 170.0334,
|
| 1367 |
+
"eval_samples_per_second": 23.525,
|
| 1368 |
+
"eval_steps_per_second": 0.371,
|
| 1369 |
+
"step": 850
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 0.086,
|
| 1373 |
+
"grad_norm": 0.21147912740707397,
|
| 1374 |
+
"learning_rate": 4.3e-05,
|
| 1375 |
+
"loss": 0.1383,
|
| 1376 |
+
"step": 860
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 0.086,
|
| 1380 |
+
"eval_cos_sim": 0.8522551655769348,
|
| 1381 |
+
"eval_loss": 0.14874456455456675,
|
| 1382 |
+
"eval_runtime": 160.1434,
|
| 1383 |
+
"eval_samples_per_second": 24.978,
|
| 1384 |
+
"eval_steps_per_second": 0.393,
|
| 1385 |
+
"step": 860
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 0.087,
|
| 1389 |
+
"grad_norm": 0.20831769704818726,
|
| 1390 |
+
"learning_rate": 4.35e-05,
|
| 1391 |
+
"loss": 0.1386,
|
| 1392 |
+
"step": 870
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 0.087,
|
| 1396 |
+
"eval_cos_sim": 0.8532180786132812,
|
| 1397 |
+
"eval_loss": 0.1477800307487196,
|
| 1398 |
+
"eval_runtime": 170.3934,
|
| 1399 |
+
"eval_samples_per_second": 23.475,
|
| 1400 |
+
"eval_steps_per_second": 0.37,
|
| 1401 |
+
"step": 870
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 0.088,
|
| 1405 |
+
"grad_norm": 0.3051137328147888,
|
| 1406 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 1407 |
+
"loss": 0.141,
|
| 1408 |
+
"step": 880
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 0.088,
|
| 1412 |
+
"eval_cos_sim": 0.8542919754981995,
|
| 1413 |
+
"eval_loss": 0.14670453121411264,
|
| 1414 |
+
"eval_runtime": 165.3769,
|
| 1415 |
+
"eval_samples_per_second": 24.187,
|
| 1416 |
+
"eval_steps_per_second": 0.381,
|
| 1417 |
+
"step": 880
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.089,
|
| 1421 |
+
"grad_norm": 0.19627386331558228,
|
| 1422 |
+
"learning_rate": 4.4500000000000004e-05,
|
| 1423 |
+
"loss": 0.1423,
|
| 1424 |
+
"step": 890
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.089,
|
| 1428 |
+
"eval_cos_sim": 0.8539001941680908,
|
| 1429 |
+
"eval_loss": 0.1470875745032972,
|
| 1430 |
+
"eval_runtime": 163.4789,
|
| 1431 |
+
"eval_samples_per_second": 24.468,
|
| 1432 |
+
"eval_steps_per_second": 0.385,
|
| 1433 |
+
"step": 890
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 0.09,
|
| 1437 |
+
"grad_norm": 0.1867413967847824,
|
| 1438 |
+
"learning_rate": 4.5e-05,
|
| 1439 |
+
"loss": 0.1371,
|
| 1440 |
+
"step": 900
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 0.09,
|
| 1444 |
+
"eval_cos_sim": 0.8544667959213257,
|
| 1445 |
+
"eval_loss": 0.1465196500037855,
|
| 1446 |
+
"eval_runtime": 159.8165,
|
| 1447 |
+
"eval_samples_per_second": 25.029,
|
| 1448 |
+
"eval_steps_per_second": 0.394,
|
| 1449 |
+
"step": 900
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 0.091,
|
| 1453 |
+
"grad_norm": 0.21700890362262726,
|
| 1454 |
+
"learning_rate": 4.55e-05,
|
| 1455 |
+
"loss": 0.1405,
|
| 1456 |
+
"step": 910
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 0.091,
|
| 1460 |
+
"eval_cos_sim": 0.8560004234313965,
|
| 1461 |
+
"eval_loss": 0.14498550846325817,
|
| 1462 |
+
"eval_runtime": 166.8423,
|
| 1463 |
+
"eval_samples_per_second": 23.975,
|
| 1464 |
+
"eval_steps_per_second": 0.378,
|
| 1465 |
+
"step": 910
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.092,
|
| 1469 |
+
"grad_norm": 0.17160160839557648,
|
| 1470 |
+
"learning_rate": 4.600000000000001e-05,
|
| 1471 |
+
"loss": 0.1387,
|
| 1472 |
+
"step": 920
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.092,
|
| 1476 |
+
"eval_cos_sim": 0.8561151623725891,
|
| 1477 |
+
"eval_loss": 0.14487657501446666,
|
| 1478 |
+
"eval_runtime": 166.4947,
|
| 1479 |
+
"eval_samples_per_second": 24.025,
|
| 1480 |
+
"eval_steps_per_second": 0.378,
|
| 1481 |
+
"step": 920
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 0.093,
|
| 1485 |
+
"grad_norm": 0.18101643025875092,
|
| 1486 |
+
"learning_rate": 4.6500000000000005e-05,
|
| 1487 |
+
"loss": 0.1362,
|
| 1488 |
+
"step": 930
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 0.093,
|
| 1492 |
+
"eval_cos_sim": 0.8565106987953186,
|
| 1493 |
+
"eval_loss": 0.1444753079627699,
|
| 1494 |
+
"eval_runtime": 162.5163,
|
| 1495 |
+
"eval_samples_per_second": 24.613,
|
| 1496 |
+
"eval_steps_per_second": 0.388,
|
| 1497 |
+
"step": 930
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 0.094,
|
| 1501 |
+
"grad_norm": 0.1913219839334488,
|
| 1502 |
+
"learning_rate": 4.7e-05,
|
| 1503 |
+
"loss": 0.137,
|
| 1504 |
+
"step": 940
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 0.094,
|
| 1508 |
+
"eval_cos_sim": 0.8572973608970642,
|
| 1509 |
+
"eval_loss": 0.14369207908856332,
|
| 1510 |
+
"eval_runtime": 159.6717,
|
| 1511 |
+
"eval_samples_per_second": 25.051,
|
| 1512 |
+
"eval_steps_per_second": 0.395,
|
| 1513 |
+
"step": 940
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 0.095,
|
| 1517 |
+
"grad_norm": 0.18858781456947327,
|
| 1518 |
+
"learning_rate": 4.75e-05,
|
| 1519 |
+
"loss": 0.1365,
|
| 1520 |
+
"step": 950
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 0.095,
|
| 1524 |
+
"eval_cos_sim": 0.8581430912017822,
|
| 1525 |
+
"eval_loss": 0.1428443865989393,
|
| 1526 |
+
"eval_runtime": 160.8669,
|
| 1527 |
+
"eval_samples_per_second": 24.865,
|
| 1528 |
+
"eval_steps_per_second": 0.392,
|
| 1529 |
+
"step": 950
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.096,
|
| 1533 |
+
"grad_norm": 0.1602417230606079,
|
| 1534 |
+
"learning_rate": 4.8e-05,
|
| 1535 |
+
"loss": 0.1353,
|
| 1536 |
+
"step": 960
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.096,
|
| 1540 |
+
"eval_cos_sim": 0.8581015467643738,
|
| 1541 |
+
"eval_loss": 0.14288127615200938,
|
| 1542 |
+
"eval_runtime": 164.3422,
|
| 1543 |
+
"eval_samples_per_second": 24.339,
|
| 1544 |
+
"eval_steps_per_second": 0.383,
|
| 1545 |
+
"step": 960
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 0.097,
|
| 1549 |
+
"grad_norm": 0.1659657508134842,
|
| 1550 |
+
"learning_rate": 4.85e-05,
|
| 1551 |
+
"loss": 0.1347,
|
| 1552 |
+
"step": 970
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 0.097,
|
| 1556 |
+
"eval_cos_sim": 0.8594202399253845,
|
| 1557 |
+
"eval_loss": 0.14155777981030407,
|
| 1558 |
+
"eval_runtime": 167.1461,
|
| 1559 |
+
"eval_samples_per_second": 23.931,
|
| 1560 |
+
"eval_steps_per_second": 0.377,
|
| 1561 |
+
"step": 970
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 0.098,
|
| 1565 |
+
"grad_norm": 0.21223703026771545,
|
| 1566 |
+
"learning_rate": 4.9e-05,
|
| 1567 |
+
"loss": 0.138,
|
| 1568 |
+
"step": 980
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 0.098,
|
| 1572 |
+
"eval_cos_sim": 0.8584538698196411,
|
| 1573 |
+
"eval_loss": 0.14252794553982676,
|
| 1574 |
+
"eval_runtime": 167.1973,
|
| 1575 |
+
"eval_samples_per_second": 23.924,
|
| 1576 |
+
"eval_steps_per_second": 0.377,
|
| 1577 |
+
"step": 980
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.099,
|
| 1581 |
+
"grad_norm": 0.15602050721645355,
|
| 1582 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 1583 |
+
"loss": 0.1392,
|
| 1584 |
+
"step": 990
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.099,
|
| 1588 |
+
"eval_cos_sim": 0.859849750995636,
|
| 1589 |
+
"eval_loss": 0.141124472162599,
|
| 1590 |
+
"eval_runtime": 168.2536,
|
| 1591 |
+
"eval_samples_per_second": 23.774,
|
| 1592 |
+
"eval_steps_per_second": 0.374,
|
| 1593 |
+
"step": 990
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 0.1,
|
| 1597 |
+
"grad_norm": 0.1410510390996933,
|
| 1598 |
+
"learning_rate": 5e-05,
|
| 1599 |
+
"loss": 0.1359,
|
| 1600 |
+
"step": 1000
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 0.1,
|
| 1604 |
+
"eval_cos_sim": 0.8603095412254333,
|
| 1605 |
+
"eval_loss": 0.14066851665722788,
|
| 1606 |
+
"eval_runtime": 162.5936,
|
| 1607 |
+
"eval_samples_per_second": 24.601,
|
| 1608 |
+
"eval_steps_per_second": 0.387,
|
| 1609 |
+
"step": 1000
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 0.101,
|
| 1613 |
+
"grad_norm": 0.15686458349227905,
|
| 1614 |
+
"learning_rate": 4.999984769144476e-05,
|
| 1615 |
+
"loss": 0.1354,
|
| 1616 |
+
"step": 1010
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 0.101,
|
| 1620 |
+
"eval_cos_sim": 0.8605126142501831,
|
| 1621 |
+
"eval_loss": 0.14045782043682994,
|
| 1622 |
+
"eval_runtime": 164.632,
|
| 1623 |
+
"eval_samples_per_second": 24.297,
|
| 1624 |
+
"eval_steps_per_second": 0.383,
|
| 1625 |
+
"step": 1010
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 0.102,
|
| 1629 |
+
"grad_norm": 0.1402912735939026,
|
| 1630 |
+
"learning_rate": 4.999939076763487e-05,
|
| 1631 |
+
"loss": 0.1333,
|
| 1632 |
+
"step": 1020
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 0.102,
|
| 1636 |
+
"eval_cos_sim": 0.8615362644195557,
|
| 1637 |
+
"eval_loss": 0.1394294276450819,
|
| 1638 |
+
"eval_runtime": 165.0933,
|
| 1639 |
+
"eval_samples_per_second": 24.229,
|
| 1640 |
+
"eval_steps_per_second": 0.382,
|
| 1641 |
+
"step": 1020
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.103,
|
| 1645 |
+
"grad_norm": 0.14782707393169403,
|
| 1646 |
+
"learning_rate": 4.999862923413781e-05,
|
| 1647 |
+
"loss": 0.1354,
|
| 1648 |
+
"step": 1030
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.103,
|
| 1652 |
+
"eval_cos_sim": 0.8612933158874512,
|
| 1653 |
+
"eval_loss": 0.13967935611950816,
|
| 1654 |
+
"eval_runtime": 164.3708,
|
| 1655 |
+
"eval_samples_per_second": 24.335,
|
| 1656 |
+
"eval_steps_per_second": 0.383,
|
| 1657 |
+
"step": 1030
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 0.104,
|
| 1661 |
+
"grad_norm": 0.14607678353786469,
|
| 1662 |
+
"learning_rate": 4.999756310023261e-05,
|
| 1663 |
+
"loss": 0.1336,
|
| 1664 |
+
"step": 1040
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 0.104,
|
| 1668 |
+
"eval_cos_sim": 0.861626148223877,
|
| 1669 |
+
"eval_loss": 0.13934949829327525,
|
| 1670 |
+
"eval_runtime": 165.2839,
|
| 1671 |
+
"eval_samples_per_second": 24.201,
|
| 1672 |
+
"eval_steps_per_second": 0.381,
|
| 1673 |
+
"step": 1040
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 0.105,
|
| 1677 |
+
"grad_norm": 0.1559648960828781,
|
| 1678 |
+
"learning_rate": 4.9996192378909786e-05,
|
| 1679 |
+
"loss": 0.1382,
|
| 1680 |
+
"step": 1050
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 0.105,
|
| 1684 |
+
"eval_cos_sim": 0.8621127009391785,
|
| 1685 |
+
"eval_loss": 0.13886528732525766,
|
| 1686 |
+
"eval_runtime": 175.2773,
|
| 1687 |
+
"eval_samples_per_second": 22.821,
|
| 1688 |
+
"eval_steps_per_second": 0.359,
|
| 1689 |
+
"step": 1050
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.106,
|
| 1693 |
+
"grad_norm": 0.15351799130439758,
|
| 1694 |
+
"learning_rate": 4.999451708687114e-05,
|
| 1695 |
+
"loss": 0.133,
|
| 1696 |
+
"step": 1060
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.106,
|
| 1700 |
+
"eval_cos_sim": 0.8621622920036316,
|
| 1701 |
+
"eval_loss": 0.1388200798247999,
|
| 1702 |
+
"eval_runtime": 168.4264,
|
| 1703 |
+
"eval_samples_per_second": 23.749,
|
| 1704 |
+
"eval_steps_per_second": 0.374,
|
| 1705 |
+
"step": 1060
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 0.107,
|
| 1709 |
+
"grad_norm": 0.13820523023605347,
|
| 1710 |
+
"learning_rate": 4.999253724452958e-05,
|
| 1711 |
+
"loss": 0.1314,
|
| 1712 |
+
"step": 1070
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 0.107,
|
| 1716 |
+
"eval_cos_sim": 0.8628634810447693,
|
| 1717 |
+
"eval_loss": 0.13812118484723032,
|
| 1718 |
+
"eval_runtime": 160.5977,
|
| 1719 |
+
"eval_samples_per_second": 24.907,
|
| 1720 |
+
"eval_steps_per_second": 0.392,
|
| 1721 |
+
"step": 1070
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 0.108,
|
| 1725 |
+
"grad_norm": 0.1627858728170395,
|
| 1726 |
+
"learning_rate": 4.999025287600886e-05,
|
| 1727 |
+
"loss": 0.1327,
|
| 1728 |
+
"step": 1080
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 0.108,
|
| 1732 |
+
"eval_cos_sim": 0.8630489706993103,
|
| 1733 |
+
"eval_loss": 0.13793376447903574,
|
| 1734 |
+
"eval_runtime": 165.8793,
|
| 1735 |
+
"eval_samples_per_second": 24.114,
|
| 1736 |
+
"eval_steps_per_second": 0.38,
|
| 1737 |
+
"step": 1080
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 0.109,
|
| 1741 |
+
"grad_norm": 0.14537328481674194,
|
| 1742 |
+
"learning_rate": 4.998766400914329e-05,
|
| 1743 |
+
"loss": 0.1347,
|
| 1744 |
+
"step": 1090
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 0.109,
|
| 1748 |
+
"eval_cos_sim": 0.8635595440864563,
|
| 1749 |
+
"eval_loss": 0.13742092325436533,
|
| 1750 |
+
"eval_runtime": 175.092,
|
| 1751 |
+
"eval_samples_per_second": 22.845,
|
| 1752 |
+
"eval_steps_per_second": 0.36,
|
| 1753 |
+
"step": 1090
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.11,
|
| 1757 |
+
"grad_norm": 0.15490968525409698,
|
| 1758 |
+
"learning_rate": 4.99847706754774e-05,
|
| 1759 |
+
"loss": 0.1322,
|
| 1760 |
+
"step": 1100
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.11,
|
| 1764 |
+
"eval_cos_sim": 0.8640041351318359,
|
| 1765 |
+
"eval_loss": 0.13697428562390268,
|
| 1766 |
+
"eval_runtime": 171.1077,
|
| 1767 |
+
"eval_samples_per_second": 23.377,
|
| 1768 |
+
"eval_steps_per_second": 0.368,
|
| 1769 |
+
"step": 1100
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 0.111,
|
| 1773 |
+
"grad_norm": 0.13356918096542358,
|
| 1774 |
+
"learning_rate": 4.998157291026553e-05,
|
| 1775 |
+
"loss": 0.1346,
|
| 1776 |
+
"step": 1110
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 0.111,
|
| 1780 |
+
"eval_cos_sim": 0.864433228969574,
|
| 1781 |
+
"eval_loss": 0.13654097273098886,
|
| 1782 |
+
"eval_runtime": 165.0965,
|
| 1783 |
+
"eval_samples_per_second": 24.228,
|
| 1784 |
+
"eval_steps_per_second": 0.382,
|
| 1785 |
+
"step": 1110
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 0.112,
|
| 1789 |
+
"grad_norm": 0.1531037539243698,
|
| 1790 |
+
"learning_rate": 4.997807075247146e-05,
|
| 1791 |
+
"loss": 0.131,
|
| 1792 |
+
"step": 1120
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 0.112,
|
| 1796 |
+
"eval_cos_sim": 0.8644208908081055,
|
| 1797 |
+
"eval_loss": 0.13655049898373545,
|
| 1798 |
+
"eval_runtime": 175.2585,
|
| 1799 |
+
"eval_samples_per_second": 22.823,
|
| 1800 |
+
"eval_steps_per_second": 0.359,
|
| 1801 |
+
"step": 1120
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.113,
|
| 1805 |
+
"grad_norm": 0.14743933081626892,
|
| 1806 |
+
"learning_rate": 4.997426424476787e-05,
|
| 1807 |
+
"loss": 0.131,
|
| 1808 |
+
"step": 1130
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.113,
|
| 1812 |
+
"eval_cos_sim": 0.8642333745956421,
|
| 1813 |
+
"eval_loss": 0.1367411518310255,
|
| 1814 |
+
"eval_runtime": 175.2554,
|
| 1815 |
+
"eval_samples_per_second": 22.824,
|
| 1816 |
+
"eval_steps_per_second": 0.359,
|
| 1817 |
+
"step": 1130
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 0.114,
|
| 1821 |
+
"grad_norm": 0.15171775221824646,
|
| 1822 |
+
"learning_rate": 4.997015343353585e-05,
|
| 1823 |
+
"loss": 0.1288,
|
| 1824 |
+
"step": 1140
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 0.114,
|
| 1828 |
+
"eval_cos_sim": 0.864815890789032,
|
| 1829 |
+
"eval_loss": 0.13617111923443737,
|
| 1830 |
+
"eval_runtime": 168.6886,
|
| 1831 |
+
"eval_samples_per_second": 23.712,
|
| 1832 |
+
"eval_steps_per_second": 0.373,
|
| 1833 |
+
"step": 1140
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 0.115,
|
| 1837 |
+
"grad_norm": 0.17678213119506836,
|
| 1838 |
+
"learning_rate": 4.996573836886435e-05,
|
| 1839 |
+
"loss": 0.1318,
|
| 1840 |
+
"step": 1150
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 0.115,
|
| 1844 |
+
"eval_cos_sim": 0.865388035774231,
|
| 1845 |
+
"eval_loss": 0.13559978248822155,
|
| 1846 |
+
"eval_runtime": 164.5713,
|
| 1847 |
+
"eval_samples_per_second": 24.306,
|
| 1848 |
+
"eval_steps_per_second": 0.383,
|
| 1849 |
+
"step": 1150
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 0.116,
|
| 1853 |
+
"grad_norm": 0.19686037302017212,
|
| 1854 |
+
"learning_rate": 4.996101910454953e-05,
|
| 1855 |
+
"loss": 0.1328,
|
| 1856 |
+
"step": 1160
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 0.116,
|
| 1860 |
+
"eval_cos_sim": 0.8653638362884521,
|
| 1861 |
+
"eval_loss": 0.13562491752850475,
|
| 1862 |
+
"eval_runtime": 162.5909,
|
| 1863 |
+
"eval_samples_per_second": 24.602,
|
| 1864 |
+
"eval_steps_per_second": 0.387,
|
| 1865 |
+
"step": 1160
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.117,
|
| 1869 |
+
"grad_norm": 0.13033832609653473,
|
| 1870 |
+
"learning_rate": 4.995599569809414e-05,
|
| 1871 |
+
"loss": 0.1314,
|
| 1872 |
+
"step": 1170
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.117,
|
| 1876 |
+
"eval_cos_sim": 0.8652835488319397,
|
| 1877 |
+
"eval_loss": 0.1357041087363905,
|
| 1878 |
+
"eval_runtime": 166.4639,
|
| 1879 |
+
"eval_samples_per_second": 24.029,
|
| 1880 |
+
"eval_steps_per_second": 0.378,
|
| 1881 |
+
"step": 1170
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 0.118,
|
| 1885 |
+
"grad_norm": 0.14916931092739105,
|
| 1886 |
+
"learning_rate": 4.995066821070679e-05,
|
| 1887 |
+
"loss": 0.1313,
|
| 1888 |
+
"step": 1180
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 0.118,
|
| 1892 |
+
"eval_cos_sim": 0.8656159043312073,
|
| 1893 |
+
"eval_loss": 0.13536836673962535,
|
| 1894 |
+
"eval_runtime": 162.1775,
|
| 1895 |
+
"eval_samples_per_second": 24.664,
|
| 1896 |
+
"eval_steps_per_second": 0.388,
|
| 1897 |
+
"step": 1180
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 0.119,
|
| 1901 |
+
"grad_norm": 0.14063435792922974,
|
| 1902 |
+
"learning_rate": 4.994503670730125e-05,
|
| 1903 |
+
"loss": 0.1292,
|
| 1904 |
+
"step": 1190
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 0.119,
|
| 1908 |
+
"eval_cos_sim": 0.8658972978591919,
|
| 1909 |
+
"eval_loss": 0.13509506275402963,
|
| 1910 |
+
"eval_runtime": 163.733,
|
| 1911 |
+
"eval_samples_per_second": 24.43,
|
| 1912 |
+
"eval_steps_per_second": 0.385,
|
| 1913 |
+
"step": 1190
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.12,
|
| 1917 |
+
"grad_norm": 0.12632516026496887,
|
| 1918 |
+
"learning_rate": 4.993910125649561e-05,
|
| 1919 |
+
"loss": 0.1291,
|
| 1920 |
+
"step": 1200
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.12,
|
| 1924 |
+
"eval_cos_sim": 0.8662503361701965,
|
| 1925 |
+
"eval_loss": 0.13473713161694467,
|
| 1926 |
+
"eval_runtime": 165.846,
|
| 1927 |
+
"eval_samples_per_second": 24.119,
|
| 1928 |
+
"eval_steps_per_second": 0.38,
|
| 1929 |
+
"step": 1200
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 0.121,
|
| 1933 |
+
"grad_norm": 0.1310836374759674,
|
| 1934 |
+
"learning_rate": 4.9932861930611454e-05,
|
| 1935 |
+
"loss": 0.1323,
|
| 1936 |
+
"step": 1210
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 0.121,
|
| 1940 |
+
"eval_cos_sim": 0.8662631511688232,
|
| 1941 |
+
"eval_loss": 0.13472085668789804,
|
| 1942 |
+
"eval_runtime": 165.8256,
|
| 1943 |
+
"eval_samples_per_second": 24.122,
|
| 1944 |
+
"eval_steps_per_second": 0.38,
|
| 1945 |
+
"step": 1210
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 0.122,
|
| 1949 |
+
"grad_norm": 0.14743076264858246,
|
| 1950 |
+
"learning_rate": 4.992631880567301e-05,
|
| 1951 |
+
"loss": 0.1296,
|
| 1952 |
+
"step": 1220
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 0.122,
|
| 1956 |
+
"eval_cos_sim": 0.8664645552635193,
|
| 1957 |
+
"eval_loss": 0.13453043796765268,
|
| 1958 |
+
"eval_runtime": 161.1688,
|
| 1959 |
+
"eval_samples_per_second": 24.819,
|
| 1960 |
+
"eval_steps_per_second": 0.391,
|
| 1961 |
+
"step": 1220
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 0.123,
|
| 1965 |
+
"grad_norm": 0.14941203594207764,
|
| 1966 |
+
"learning_rate": 4.991947196140618e-05,
|
| 1967 |
+
"loss": 0.127,
|
| 1968 |
+
"step": 1230
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 0.123,
|
| 1972 |
+
"eval_cos_sim": 0.8669182658195496,
|
| 1973 |
+
"eval_loss": 0.1340716586326307,
|
| 1974 |
+
"eval_runtime": 159.0609,
|
| 1975 |
+
"eval_samples_per_second": 25.148,
|
| 1976 |
+
"eval_steps_per_second": 0.396,
|
| 1977 |
+
"step": 1230
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.124,
|
| 1981 |
+
"grad_norm": 0.15963555872440338,
|
| 1982 |
+
"learning_rate": 4.991232148123761e-05,
|
| 1983 |
+
"loss": 0.1286,
|
| 1984 |
+
"step": 1240
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.124,
|
| 1988 |
+
"eval_cos_sim": 0.8672899007797241,
|
| 1989 |
+
"eval_loss": 0.13370788147198617,
|
| 1990 |
+
"eval_runtime": 162.1984,
|
| 1991 |
+
"eval_samples_per_second": 24.661,
|
| 1992 |
+
"eval_steps_per_second": 0.388,
|
| 1993 |
+
"step": 1240
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 0.125,
|
| 1997 |
+
"grad_norm": 0.12001099437475204,
|
| 1998 |
+
"learning_rate": 4.990486745229364e-05,
|
| 1999 |
+
"loss": 0.1289,
|
| 2000 |
+
"step": 1250
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 0.125,
|
| 2004 |
+
"eval_cos_sim": 0.8670727610588074,
|
| 2005 |
+
"eval_loss": 0.13392866852032603,
|
| 2006 |
+
"eval_runtime": 161.0006,
|
| 2007 |
+
"eval_samples_per_second": 24.845,
|
| 2008 |
+
"eval_steps_per_second": 0.391,
|
| 2009 |
+
"step": 1250
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 0.126,
|
| 2013 |
+
"grad_norm": 0.13203372061252594,
|
| 2014 |
+
"learning_rate": 4.989710996539926e-05,
|
| 2015 |
+
"loss": 0.1268,
|
| 2016 |
+
"step": 1260
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 0.126,
|
| 2020 |
+
"eval_cos_sim": 0.8671303987503052,
|
| 2021 |
+
"eval_loss": 0.1338693347190565,
|
| 2022 |
+
"eval_runtime": 164.8756,
|
| 2023 |
+
"eval_samples_per_second": 24.261,
|
| 2024 |
+
"eval_steps_per_second": 0.382,
|
| 2025 |
+
"step": 1260
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.127,
|
| 2029 |
+
"grad_norm": 0.1145872101187706,
|
| 2030 |
+
"learning_rate": 4.9889049115077005e-05,
|
| 2031 |
+
"loss": 0.1274,
|
| 2032 |
+
"step": 1270
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.127,
|
| 2036 |
+
"eval_cos_sim": 0.8675019145011902,
|
| 2037 |
+
"eval_loss": 0.13349107553708017,
|
| 2038 |
+
"eval_runtime": 169.3092,
|
| 2039 |
+
"eval_samples_per_second": 23.625,
|
| 2040 |
+
"eval_steps_per_second": 0.372,
|
| 2041 |
+
"step": 1270
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 0.128,
|
| 2045 |
+
"grad_norm": 0.1377398520708084,
|
| 2046 |
+
"learning_rate": 4.988068499954578e-05,
|
| 2047 |
+
"loss": 0.1317,
|
| 2048 |
+
"step": 1280
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 0.128,
|
| 2052 |
+
"eval_cos_sim": 0.8676916360855103,
|
| 2053 |
+
"eval_loss": 0.1333048849319166,
|
| 2054 |
+
"eval_runtime": 159.5371,
|
| 2055 |
+
"eval_samples_per_second": 25.073,
|
| 2056 |
+
"eval_steps_per_second": 0.395,
|
| 2057 |
+
"step": 1280
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 0.129,
|
| 2061 |
+
"grad_norm": 0.170301154255867,
|
| 2062 |
+
"learning_rate": 4.987201772071971e-05,
|
| 2063 |
+
"loss": 0.1285,
|
| 2064 |
+
"step": 1290
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 0.129,
|
| 2068 |
+
"eval_cos_sim": 0.8680738806724548,
|
| 2069 |
+
"eval_loss": 0.13292226507412852,
|
| 2070 |
+
"eval_runtime": 176.1386,
|
| 2071 |
+
"eval_samples_per_second": 22.709,
|
| 2072 |
+
"eval_steps_per_second": 0.358,
|
| 2073 |
+
"step": 1290
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 0.13,
|
| 2077 |
+
"grad_norm": 0.1578381359577179,
|
| 2078 |
+
"learning_rate": 4.9863047384206835e-05,
|
| 2079 |
+
"loss": 0.1295,
|
| 2080 |
+
"step": 1300
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 0.13,
|
| 2084 |
+
"eval_cos_sim": 0.8683300614356995,
|
| 2085 |
+
"eval_loss": 0.13265489532696664,
|
| 2086 |
+
"eval_runtime": 164.7229,
|
| 2087 |
+
"eval_samples_per_second": 24.283,
|
| 2088 |
+
"eval_steps_per_second": 0.382,
|
| 2089 |
+
"step": 1300
|
| 2090 |
+
}
|
| 2091 |
+
],
|
| 2092 |
+
"logging_steps": 10,
|
| 2093 |
+
"max_steps": 10000,
|
| 2094 |
+
"num_input_tokens_seen": 0,
|
| 2095 |
+
"num_train_epochs": 9223372036854775807,
|
| 2096 |
+
"save_steps": 10,
|
| 2097 |
+
"stateful_callbacks": {
|
| 2098 |
+
"TrainerControl": {
|
| 2099 |
+
"args": {
|
| 2100 |
+
"should_epoch_stop": false,
|
| 2101 |
+
"should_evaluate": false,
|
| 2102 |
+
"should_log": false,
|
| 2103 |
+
"should_save": true,
|
| 2104 |
+
"should_training_stop": false
|
| 2105 |
+
},
|
| 2106 |
+
"attributes": {}
|
| 2107 |
+
}
|
| 2108 |
+
},
|
| 2109 |
+
"total_flos": 0.0,
|
| 2110 |
+
"train_batch_size": 160,
|
| 2111 |
+
"trial_name": null,
|
| 2112 |
+
"trial_params": null
|
| 2113 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:228943860476aef422f429bf676bbb7f95e79b4b2c6e641a22176011b8628860
|
| 3 |
+
size 5176
|